Live  /  DigitalOcean  /  Incident

DigitalOcean: Serverless Inference - Gemma4 Latency Issues Causing Timeouts & Slow Responses

DigitalOcean · digitalocean.comMinor incidentResolved2026-07-15 21:46 UTC
Public — what happened

Summary

DigitalOcean said that starting on July 13, 2026 at 19:04 UTC, customers using the Gemma 4 model on its Serverless Inference and Dedicated Inference platforms could see very slow responses or requests that timed out. The company attributed the problem to a backend configuration issue and reported deploying a fix that evening at 22:24 UTC, but the slowdowns returned on July 14 at 11:07 UTC, this time described as affecting Serverless Inference. DigitalOcean said a second fix went out on July 15 at 10:06 UTC, and it marked the incident resolved at 21:46 UTC that day, saying Gemma 4 response times were back to normal.

Affected

Inference

As listed on DigitalOcean’s status page.

Timeline

Source of record: provider status page. Summary written by SquidGrid from the provider’s published updates; times in UTC.

Independent path health

The status page is the symptom. SquidGrid measures the paths that reach DigitalOcean against their own baselines — whether the routes carrying your traffic degraded, at which handoff, and who owns that stretch.

🔒 Sign in to see
Customers get the live Provider Assessment Report for DigitalOcean: paths off baseline right now, where on the route conditions change, and its full incident track record.

Included with any paid plan.

Open the DigitalOcean report →

More from DigitalOcean

The full archive is on the DigitalOcean status & history page.

Related: DigitalOcean status & history. SquidGrid is not affiliated with DigitalOcean.